<<

Phylogenetic relationships of the based on nuc-25S-rDNA, mit-12S-rDNA, and mit-atp6-DNA combined sequences

a b Admir J. GIACHINl ,*, Kentaro HOSAKA , Eduardo NOUHRAc, d a Joseph SPATAFORA , James M. TRAPPE

ARTICLE INFO ABSTRACT

Phylogenetic relationships among , Gomphales, , and were estimated via combined sequences: nuclear large subunit ribosomal DNA (nuc-25S- rDNA), mitochondrial small subunit ribosomal DNA (mit-12S-rDNA), and mitochondrial atp6 DNA (mit-atp6-DNA). Eighty-one taxa comprising 19 genera and 58 were inves- tigated, including members of the Clathraceae, Gautieriaceae, Geastraceae, , , , Protophallaceae, and Sphaerobolaceae. Although some nodes Keywords: deep in the tree could not be fully resolved, some well-supported lineages were recovered, atp6 and the interrelationships among , , , and Turbinel- Gomphales Ius, and the placement of are better understood. Both Gomphus sensu lato and Rama- Homobasidiomycetes ria sensu lato comprise paraphyletic lineages within the Gomphaceae. Relationships of the rDNA subgenera of Ramaria sensu lato to each other and to other members of the Gomphales were Systematics clarified. Within Gomphus sensu lato, Gomphus sensu stricto, , Gloeocantharellus and Phaeoclavulina are separated by the presence/absence of clamp connections, orna- mentation (echinulate, verrucose, subreticulate or reticulate), and basidiomal morphology (fan-shaped, funnel-shaped or ramarioid). , a sequestrate in theGautieria- ceae, was recovered as monophyletic and nested with members of Ramaria subgenus Ramaria. This agrees with previous observations of traits shared by these two ectomycor- rhizal taxa, such as the presence of fungal mats in the soil. was recovered as a sister group to , , and . The results reaffirm relationships be- tween the Geastrales, Gomphales, Hysterangiales, and the Phallales, suggesting extensive convergence in basidiomal morphology among members of these groups. A more extensive sampling that focuses on other loci (protein-coding genes have been shown to be phyloge- netically informative) may be useful to answer questions about evolutionary relationships among these fungal groups. © 2010 The British Mycological Society. Published by Elsevier Ltd. All rights reserved. (1954), Heinemann (1958), Donk (1964),Giachini (2004) and Introduction Hosaka et al. (2006), Clavariadelphus is a member of the Gom- phales. The results of Villegas et al. (1999), however, disagree The gomphoid fungi occupy a unique position in the phylog- with the premises that Clavariadelphus, Gomphus, and Ramaria eny of higher Basidiomycetes (Bruns et al. 1998; Pine et al. are members of the same order. According to those authors 1999; Hibbett & Thorn 2001; Hosaka et al. 2006). They are prom- Clavariadelphus is not grouped within but rather a sister group inent in most forest ecosystems as saprotrophs and mutual- to the Gomphales. ists. The fungi in this group are also characterized by a wide Pine et al. (1999) and Humpert et al. (2001), on the other range of basidiomal morphologies, from stalked ramarioid/ hand, using sequences of both mitochondrial (mit-12S- clavarioid to cantharelloid-gomphoid, clavate, resupinate- rDNA) and nuclear (nuc-25S-rDNA) loci, showed that a gom- odontoid, to sequestrate. Molecular studies reveal that gom- phoid-phalloid clade including Clavariadelphus, , phoid fungi are closely related to taxa in the Geastrales, Hys- Gloeocantharellus, Gomphus (monophyletic), Lentaria, Ramaria terangiales, and Phallales (Colgan et al. 1997; Hibbett et al. (paraphyletic), , and was recovered 1997; Pine et al. 1999; Humpert et al. 2001; Hosaka et al. 2006). in all analyses performed. Pine et al. (1999) showed that Gom- of the Gomphales has traditionally relied upon phus sensu lato represented a terminal monophyletic group in morphological characters now known to be subject to parallel the gomphoid-phalloid clade (although just two taxa were evolution and phenotypic plasticity (Moncalvo et al. 2000). sampled), having Ramaria as sister group (Figs 1-3 in Pine Consequently, many current genera and families are artificial, et al. 1999). Based on morphological as well as molecular and taxonomic limits and identity of natural groups in the or- characters, Giachini (2004) revisited the generic concepts in ders Geastrales, Gomphales, Hysterangiales, and Phallales are the family Gomphaceae and recombined the species being re-examined. of Gomphus sensu lato into Gloeocantharellus, Gomphus Past workers (Maire 1902, 1914; Eriksson 1954; Heim 1954) sensu stricto, and the resurrected genera Phaeoclavulina and recognized the relatedness of diverse morphologies within Turbinellus. the Gomphales from microscopic and macrochemical charac- Cantharelloid/gomphoid and clavarioid fungi have his- ters, including cyanophilic spore ornamentation, chiastic torically been prominent in hypotheses about the origin of basidia, hyphal construction, and positive hymenial reaction fleshy basidiomycetes (Singer 1947, 1986; Heim 1954; to ferric sulfate (Eriksson 1954; Donk 1961, 1964; Petersen Corner 1966;Harrison 1971; Petersen 1971a; Corner 1972; 1971a; Villegas et al. 1999).Donk (1961,1964)proposed the fam- julich 1981; Miller & Watling 1987). Their fruiting forms can ily Gomphaceae to include the resupinate-odontoid genera be arranged in a transformation series, from clavate at one Kavinia and , the stalked clavarioid-ramarioid gen- end, cantharelloid/gomphoid intermediately, and agaricoid era Lentaria and Ramaria, the stalked hydnoid genus Beenakia, at the other extreme. Corner (1972) proposed the "Clava ria the stipitate agaricoid genus Gloeocantharellus, and the pile ate theory" of basidiomycete evolution in which cantharelloid genera Chloroneuron and Gomphus. Corner (1970) proposed and clavarioid fungi were to be regarded as ancestral, and Ramariaceae to include Delentaria, Kavinia, Lentaria, and Rama- from which all other Homobasidiomycetes have been de- ria. He excluded the pileate genera because no intermediate rived. He suggested that simple clavate morphologies (e.g. species linked the gomphoid and ramarioid morphologies. Clavaria) with smooth hymenia gave rise to intermediate Petersen (1971a) suggested a gomphoid ancestral morphology cantharelloid species (e.g. , ), and for the family and later revised Donk's and Corner's familial from those were derived the wrinkled or folded hymenial classifications to include Beenakia, Gomphus, Kavinia, Ramari- gomphoid species (e.g. Gomphus, Turbinellus). Other authors cium, , and Ramaria (Petersen 1973, 1988). agree on transformations among ramarioid, cantharelloid, Morphological and recent molecular data (mitochondrial and agaricoid forms but propose the opposite polarity, sug- and nuclear rONA) have been used to infer inter- and intra- gesting that lineages containing cantharelloid, ramarioid, specific relationships among genera in this group of fungi. and club-like fungi have been derived from agaricoid ances- According to Pine et al. (1999), Villegas et al. (1999), Humpert tors (Fiasson et al. 1970; Arpin & Fiasson 1971; Petersen et al. (2001), and Hosaka et al. (2006), the Gomphales includes 1971a; Singer 1986). the genera Beenakia, Clavariadelphus, Gautieria, Gloeocantharel- In this paper we analyze phylogenetic relationships among Ius, Gomphus, Kavinia, Lentaria, Phaeoclavulina, Ramaria, Ramari- major evolutionary lineages of gomphoid fungi using com- cium, and Turbinellus. Hosaka et al. (2006) demonstrated the bined sequence data from nuclear (nuc-25S-rDNA) and Gomphales to be a sister group to the Phallales, represented mitochondrial-encoded ribosomal and non-ribosomal RNA by the families Clathraceae (sensu Chevallier), Phallaceae genes (mit-12S-rDNA, mit-atp6-DNA). Our taxonomic sam- (sensu Corda), Lysuraceae, Protophallaceae (sensu Zeller), pling focused on the Gomphales sensu Julich (1981). Major Claustulaceae, and Trappeaceae. Both Gomphales and Phal- questions tested in this study were: lales are closely related to the Hysterangiales (sensu Hosaka & Castellano) and the Geastrales. 1) Is Gomphus sensu lato monophyletic? Villegas et al. (1999), using morphological traits, proposed 2) Are genera within Gomphus sensu lato monophyletic? the Gomphales to be monophyletic and delimited by the pres- 3) Are the Gomphales, Hysterangiales, Phallales, and Geas- ence of mycelial cords or rhizomorphs. It included the families trales closely related? Beenakiaceae (Beenakia, Kavinia, and Ramaricium), Gompha- 4) How have basidiomatal morphology, presence or absence ceae (Gomphus and Gloeocantharellus), (Lentaria), of clamp connections, and substrate affinity evolved and Ramariaceae (Ramaria). According to Singer (1949), Heim within the Gomphales? the Kishino-Hasegawa (K-H)test implementing the likelihood Materials and methods model described above (Table 2). Character state reconstruc- tion for substrate affinity, basidiomata morphology, and pres- Taxonomic sampling ence/absence of clamp connections was performed in Mr 3 Bayes (MC ) version 3.0 (Huelsenbeck & Ronquist 2001); equal The sampling of Geastrales, Gomphales, Hysterangiales, and weights for all character state transformations were assumed. Phallales included 19 genera and 58 species (total of 81 taxa) as listed along with GenBank accession numbers (Table 1). Character mapping One species each of Bondarzewia and Russula plus Cortinarius iodes were included as outgroups. Holotypes and representa- For this study, the GTR (likelihood) model of character evolu- tive specimens were examined and sampled when available. tion fit our data best. Adopting GTR with Multistate ver. 0.8 Dried specimens were obtained from the following herbaria: (Pagel 2003),we calculated trait evolution for characters repre- BPI, BR, DSH,FH, K, MICH, NYS, 0, OSA, OSC, PERTH,PDD, senting substrate affinity, presence/absence of clamp connec- SFSU,SUC, TENN,UC, and UPS(http://www.nybg.org/bsci/ tions, and basidiomatal morphology for the node/clade ih/ih.html). Two or more fresh collections of each species or denoting the order Gomphales. variety were included when available. Species identification and nomenclature were based on ho- lotype and para type specimens and taxonomic keys and spe- Results cies descriptions (Corner 1950, 1966, 1969, 1970; Petersen 1971b, 1981, 1988; Marr & Stuntz 1973; Schild 1998; Roberts Sequence alignment and nucleotide sequence variation 1999; Giachini 2004).

Alignment over a broad taxonomic sampling (84taxa) was not DNA extraction, amplification, and sequencing attainable for a few hypervariable, indel-rich regions: those were removed from the analyses. A few remaining single- DNA sequence data were obtained from three independent gap regions occurring in only one or few sequences were re- loci: LROR-LR3 region for nuclear large subunit ribosomal moved due to the possibility that they represented sequencing DNA (nuc-LSU-rDNA), MS1-MS2 region for mitochondrial errors. In contrast, several gap regions with short indels were small subunit ribosomal DNA (mt-SSU-rDNA), and ATPase recorded as phylogenetic ally informative. In all, 314 positions subunit 6 (atp6). The primers and PCRprotocols have been de- corresponding to regions with problematic alignments were scribed previously (summarized in Assembling the Fungal removed, and 120 indel positions were recorded. After re- Tree of Life website; http://aftol.org/primers.php). moval of the 5' and 3' positions (incomplete for several taxa), Phylogenetic analyses were conducted for the concate- 2063 positions remained in the final analyses. Of these, 758 nated three-locus dataset under Bayesian and parsimony cri- were constant, 328 variable characters were parsimony- teria. Maximum parsimony analyses were conducted by uninformative, and 685 were parsimony-informative. The fi- PAUP*4.0bl0 (Swofford 2002), with nodal supports tested by nal alignment is available in the web as a NEXUSfile (SN1858). bootstrap analysis. Analyses were conducted with 10 000 ran- dom additions of heuristic search with TBRand Multrees op- tion on. All MPTs recovered were subsequently compared to Phylogenetic trees each other under the maximum likelihood criterion (Kishino & Hasegawa 1989). Significant topological differences under The analyses of the three combined loci yielded three MPTs of the maximum parsimony criterion and the combinability of 4967 steps (Fig1).For those trees, the CIwas 0.287, the RI0.566, the data were estimated via the Shimodaira & Hasegawa and the RC0.163. Fig2 depicts the consensus tree obtained for (1999) likelihood test (p< 0.05). Bayesian analysis was con- the taxa studied. For this tree, Bayesian posterior probability 3 ducted by use of MrBayes ver. 3.0b4 (Huelsenbeck and Ron- values (MC ) are presented above branches and bootstrap quist 2001), with 3 000 000 generations of MCMCMC. Every values greater than 50 below branches (consensus and Bayes- 100th tree was sampled, to produce 30000 trees. Four chains ian produced identical trees). The results support the mono- were applied (one cold and three heated; temperature set to phyletic status and the close evolutionary relationship of the the default value of 0.2). We applied independent models for orders Geastrales, Gomphales, Hysterangiales and Phallales, each partition using general time reversible (GTR) and gamma corroborating results previously obtained by Hibbett et al. (d) distribution (burn-in period of 15000 trees). Stationarity (1997), Hosaka et al. (2006), Humpert et al. (2001), and Pine was determined when chains reached the arithmetic mean et al. (1999). These results agree with the findings of Giachini likelihood value of -94 331.77. (2004) on the paraphyletic status of Gomphus sensu lato and Alternative phylogenetic hypotheses reflecting different confirm the monophyletic status of Gloeocantharellus, Gom- classifications and species relationships were constructed in phus, Phaeoclavulina, and Turbinellus. Furthermore, the phylo- MacClade version 3.03 (Maddison & Maddison 1992) and genetic analyses corroborate previous results on the PAUP*version 4.0bl0 (Swofford 2002). These trees were used para phyletic status of Clavariadelphus, Kavinia, and Ramaria, as constraint starting topologies in maximum parsimony and the monophyletic status of Gautieria, Lentaria, and Ramaria analyses in the heuristic search option (100 random sequence subgenus Ramaria (Humpert et al. 2001). In the Geastrales they additions, TBR, and MULPARS off). Most parsimonious trees support a monophyletic Sphaerobolus (Fig 1), a result corrobo- recovered with and without constraints were compared by rated by Hosaka et al. (2006).

Geastrales, Gomphales, Hysterangiales, and Phallales placement of the three genera sampled for the Hysterangiales (, , and Protuber)indicates a close relation- The results of the three combined loci support the hypothesis ship of the Hysterangiales to the Gomphales (Fig 1). Hysteran- of a Geastrales-Hysterangiales-Phallales relationship to the gium, represented by North American species, was recovered Gomphales (Colgan et al. 1997; Hibbett et al. 1997; Humpert as monophyletic, disagreeing with the work of Hosaka et al. et al. 2001; Giachini 2004; Hosaka et al. 2006). Both Bayesian (2006).Our study, however, sampled a much smaller portion MC3 and bootstrap values indicate a consistent and confident of the order, as well as only three loci compared to five of resolution for the evolutionary placement of Geastrales, Hys- Hosaka et al. (2006); accordingly we accept their conclusion terangiales, and Phallales in relation to the Gomphales. The that Hysterangium is paraphyletic. We sampled four genera of the Phallales: CIathrus, , , and PseudocoIus. This Character mapping order is shown as a sister group to the Hysterangiales (Fig 1). The Geastrales, represented in this study by Geastrum (earth- Characters for substrate affinity, presence/absence of clamp star ) and SphaeroboIus (cannon-ball fungus), was recov- connections, and basidioma morphology are shown in Fig 1. ered as a basal, more primitive sister lineage (Fig 1). The Evolution of substrate affinity indicates both lignicolous and placement of earth-star and cannon-ball fungi as an ancestral terricolous substrate affinities as ancestral for the Gomphales lineage for the Gomphales-Hysterangiales-PhallaIes has been with one derivation of the strictly terricolous condition, for shown by Hosaka et al. (2006),but beyond the basic hymenomy- CIavariadeIphus and the clade containing GIoeocantharellus, cete features shared by these taxa, no other morphological Gomphus, Ramaria subgenera LaeticoIoraand Ramaria, TurbineI- characters have been identified to support this evolutionary Ius, and the sequestrate genus Gautieria (Fig 1). Evolution of line. Further sampling of the Geastrales might aid understand- basidioma macro morphology suggests that the ramarioid ing of this evolutionary placement. morphology is ancestral for the Gomphales with multiple derivations of diverse basidiomata morphologies, i.e., club, provides a different resolution than the nuc-25S-rDNA or gomphoid, odontoid, resupinate, and sequestrate (Fig 1). The mit-atp6-DNA. Since the results for the mit-12S-rDNA locus mapping of presence or absence of clamp connections shows were marginal for some combination sets, we combined the multiple gains and losses of this feature throughout the evolu- three loci for further analyses. tionary history of the Gomphales (Fig 1). The results suggest an ancestral condition with clamp connections and multiple Discussion losses of this feature occurring in certain species of Beenakia, Gautieria, Gloeocantharellus,Kauinia, Ramaria subgenera Laetico- Support for a Geastrales-Gomphales-Hysterangiales- lora and Ramaria, and Turbinellus. Phallales relationship

Kishino-Hasegawa and Shimodaira-Hasegawa tests The three independent loci examined support a close phyloge- netic relationship among the Geastrales, Gomphales, Hysteran- The results of the Kishino-Hasegawa and Shimodaira-Hase- giales, and the Phallales (Figs 1 and 2). Relationships among gawa tests are presented in Table 2 and Fig3, respectively.Hy- stinkhorns, earth-stars, the cannon-ball fungus, ramarioid-dav- potheses for the Kishino-Hasegawa test were (i) Gomphus arioid, and cantharelloid-gomphoid fungi have only recently sensu lato constrained (forced) to monophyly; and (ii) mono- been proposed in the literature (Hibbett et al. 1997; Pine et al. phyly of combination sets involving at least two ofthe genera 1999;Humpert et al. 2001; Hosaka et al. 2006).Evolutionary rela- (at the time) Gloeocantharellus, Gomphus, Phaeoclauulina and tionships for the fungi belonging to some of those groups have Turbinellus on all possible fashions. been proposed in the past, and the literature on some, i.e. the The Shimodaira-Hasegawa test was employed to test the Gomphales, has been extensively reviewed, mostly in reference combinability of the three loci sampled.Although indepen- to morphological characters (Maire 1902, 1914; Donk 1964; dent analyses of the three data sets produced somewhat dif- Corner 1966, 1969;Petersen 1968, 1971a; Giachini et al. 2001).As ferent resolutions regarding a few weakly supported for relationships among Geastrales, Gomphales, Hysterangiales, terminal nodes, the overall topological reconstructions and Phallales, identified by Pine et al. (1999)and corroborated by obtained for each of the three independent loci were similar. Humpert et al. (2001)and Hosaka et al. (2006),no unifying morpho- The Shimodaira-Hasegawa test for data combinability logical synapomorphies have been identified. showed that the nuc-25S-rDNA and mit-atp6-DNA are statisti- cally combinable at p = 0.111-0.574 (Fig 3).Mit-12S-rDNA, on the other hand, seems to be the most divergent locus, not sta- Rejection of a monophyletic Gornphus sensu lato tistically combinable with either nuc-25S-rDNA (p< 0.0001) or mit-atp6-DNA (p < 0.0001). When forcing the data from mit- Phylogenetic analyses of the combined nuc-25S-rDNA, mit- 12S-rDNA into the topology of the best parsimonious tree 12S-rDNA, and mit-atp6-DNA rejected the monophyletic con- obtained with the nuc-25S-rDNA data, the results indicate dition for Gomphus sensu lato (Table 2). The results indicate they are combinable (p = 0.001-0.751)(Fig 3). The divergence Gloeocantharellus, Gomphus, Phaeoclauulina, and Turbinellus to observed for the mit-12S-rDNA locus may be due to the faster be monophyletic genera (Table 2, Fig 1). Gautieria and Ramaria or slower rate of evolution suspected for this genomic region nested within Gomphus sensu lato, whereas Beenakia, Clauaria- when compared to the nuc-25S-rDNA and mit-atp6-DNA re- delphus, Kauinia, and Lentaria were recovered as a sister group gions. The presence of large indeIs observed throughout the to it. Clades represented by these genera received high likeli- 3 mit-12S-rDNA locus indicates that this genomic region hood (Bayesian posterior probability - MC ) and bootstrap sup- port (Fig2). The Kishino-Hasegawa test indicated significantly worse trees when Gomphus sensu lato or combinations of taxa within Gomphus sensu lato were constrained to the monophy- letic condition (Table 2), indicating that a confident resolution of the phylogeny of the Gomphales was obtained after the combination of more than one locus. This suggests that a com- bination of fast evolving (mit-12S) and protein-coding (atp6) genes provided a consistent resolution for the placement and evolutionary history of this group.

Evolution of substrate affinity

The analyses indicated an ambiguous ancestral substrate affinity condition for the Gomphales (Fig1).The most basal lineage of the Gomphales is composed of the lignicolous/terricolous genus Phaeoclauulina (Fig1).Even though some taxa, including Gautieria, Gomphus, Hysterangium,Ramaria, and Turbinellus,are known my- corrhizal associates (Masui 1926, 1927;Castellano 1988;Miller & Miller 1988;Griffiths et al. 1991;Agerer et al. 1996a, b, c, d; Agerer et al. 1998),the mycorrhizal status of most lignicolous/terricolous species of the orders treated here is still unknown. Euolution of clamp connections those of Ramaria subgenus Ramaria are ballistosporic (Humpert et al. 2001). Other cases are known where epigeous The presence of clamp connections is ancestral in the Gom- and sequestrate taxa differ in the trait of spore symmetry but phales with multiple losses of the clamped condition (Fig 1). retain other characteristics such as size, shape and ornamen- This is consistent with Corner's (1966) evolution hypothesis tation (Thiers & Trappe 1969; Thiers 1984;Bruns & Szaro for clamp connection in the Gomphales that suggests varying 1992; Mueller & Pine 1994; Hibbett et al. 1997; Lebel 1998). degrees of presence of clamp connections and assumes the "clarnpless" state as derived from the "clamped state." Humpert et al. (2001), however, showed species of Ramaria Conclusion subgenus Lentoramaria without clamp connections to be evo- lutionarily more basal. Production of clamp connections Our data corroborate previous work rejecting the monophyly of Gomphus sensu lato (Giachini 2004; Hosaka et al. 2006). Fur- varies among species of Beenakia, Kauinia, Ramaria subgenera thermore, they support the monophyletic status of Gloeocan- Laeticolora and Ramaria, and within families of the Geastrales, tharellus, Gomphus, Phaeoclauulina, and Turbinellus (Fig 1). Hysterangiales and Phallales. Invariable presence of clamp They also reinforce previous observations (Humpert et al. connections is only observed for species of Clauariadelphus, 2001; Hosaka et al. 2006) on the ramarioid ancestral condition Gomphus, Lentaria, and Phaeoclauulina. for the Gomphales, and the independent derivations of cla- vate, gomphoid, odontoid, resupinate, and sequestrate mor- Polarity of basidioma morphology phologies. The coralloid-ramarioid morphology however, Our data corroborate the hypothesis of Humpert et al. (2001) does not indicate a natural monophyletic group. Rather, this and Hosaka et al. (2006) of a ramarioid morphology as ances- condition was gained and lost several times during the course of evolution. These data also suggest an ambiguous character tral for the Gomphales, with multiple derivations of distinct basidiomatal morphologies, i.e., clavate, odontoid, gomphoid, state condition for substrate affinity in the Gomphales. Fungi within the Geastrales, Gomphales, Hysterangiales resupinate, and sequestrate (Fig 1). Petersen (1971a) hypothe- and Phallales have been known for years to have important sized the resupinate-odontoid genera Kauinia and Ramaricium ecological roles (symbionts, decomposers, etc.). Although to be derived from Ramaria subgenus Lentoramaria. Our results new information has been added, comparative studies on are consistent with the hypothesis that Beenakia, Kauinia, and Lentaria are derived from within the clavate, terricolous genus the anatomy and biochemistry of these taxa are still required to fully unveil the morphological features (synapomorphies) Clauariadelphus. that unite these fungi. Euolution of sequestrate fungi from epigeous Gomphales

Gautieria is a sequestrate genus (Gautieriaceae) in the Gautier- Acknowledgments iales (Zeller & Dodge 1918). As shown by Humpert et al. (2001) and Hosaka et al. (2006), and corroborated here, Gautieria is This research was supported in part by the Forest Mycology closely related to Ramaria, more specifically a sister group to Team, U.S. Forest Service Pacific Northwest Research Station Ramaria subgenus Ramaria, and therefore a member of the (Forest Service and USDA,Corvallis). The senior author thanks Gomphales (Fig 1). the Conselho Nacional de Desenvolvimento Cientifico e Tecno- As discussed by Fischer (1933)and Cunningham (1942),Gau- 16gico(CNPq)of the Brazilian Ministry of Education for the doc- tieria is regarded as coralloid/ramarioid with tramal plates toral fellowship. Special thanks to the following collections for growing outward from a central sterile base, resulting in the loans: BPI,BR,CANB,CUP, DAOM,DAR,F, FH, GH, H, K, L, LPS, formation of pockets or locules. Alternatively, the develop- M, MICH,NCU,NY,NYS,OULU,OSA,OSC,PC,PDD,PERTH,PH, ment ofGautieria has also been described as forate by other au- S, SFSU,TENN, TNS, UC, UPS, WTU, and ZT. Richard Halse of thors, with formation of the branches occurring from inward the Department of Botany, Oregon State University, un stint- growth of exterior tissue (Fitzpatrick 1913; Dring 1973; Miller ingly arranged loans from herbaria. Andy Taylor provided & Miller 1988).The ridged of Gautieria were once thought samples of G. clavatus from Sweden, and Leif Ryvarden for to indicate a close evolutionary relationship with the specimens of Beenakia and Kauinia. Thanks to Gi-Ho Sung and (Smith 1973). Because Gautieria forms a true hymenium arising Sarah R.Dubrasich for editorial comments on this manuscript. from the trama with its basidium tips exposed to an open chamber, it has also been suggested to have evolved from a hymenomycete (Dring 1973),more specifically from a ramar- REFERENCES ioid ancestor (Humpert et al. 2001). This finding supports the conclusion of Bruns et al. (1998), who showed that Gautieria was closely related to Gomphus, Kauinia, and Ramaria. It is also Agerer R, 1996a.Ramariaaurea (Schaeff.:Fr.)Quel.+ Fagus sylvatica consistent with the findings of Agerer (1999), that the Gom- L. Descriptions of Ectomyccrrhizue1:107-112. phales, Geastrales, and Gautieriales all have a unique, ampu- Agerer R, 1996b. Ramaria largentii Man & D.E.Stuntz + Piceaabies (L.)Karst. Descriptionsof Ectomycorrhizae1: 113-118. late rhizomorphic morphology in common: they share the Agerer R, 1996c.Ramaria spinulosa (Fr.)Que!.+ Fagus syluatica L. ramarioid, not the boletoid type. In addition, species of Gautie- Descriptions of Ectomycorrhizae 1: 119-124. ria and Ramaria subgenus Ramaria have similar ridged spores. Agerer R,1996d.Ramariasubbotrytis (Coker)Corner + Quercus robur However, the spores of Gautieria are statismosporic, whereas L.Descriptions of Ectomycorrhizae 1: 125-130. Agerer R, 1999. Never change functionally successful principle: Heim R, 1954. A propos de trois chanterelles americaines. Revue the evolution of Boletales s.l. (Hymenomycetes,Basidiomy- Mycologique 19: 47-56. cota) as seen from below-ground features. Sendtnera 6: 5-91. Heinemann P, 1958. Champignons recoltes au Congo Beige par Agerer R, Beenken L, Christian], 1998. (Pers.: Fr.) Madame M. Goosens-Fontana. III. Cantharellineae. Bulletin du S.F. Gray + Picea abies (L.) Karst. In: Agerer R, Danielson RM, Jardin botanique de l'etat, Bruxelles 28: 385-438. Egli S, Ingleby K, Luoma D, Treu R (eds), Descriptions of Hibbett DS, Pine EM, Langer E, Langer G, Donoghue M], 1997. Ectomycorrhizae 3: 25-29. Evolution of gilled and inferred from Arpin N, Fiasson ]-L, 1971. The pigments of Basidiomycetes: their ribosomal DNA sequences. Proceedings of the National Academy chemotaxonomic interest. In: Petersen RH (ed), Evolution in the of Science USA 94: 12002-12006. Higher Basidiomycetes: an International Symposium. University of Hibbett DS, Thorn RG, 2001. : Homobasidiomy- Tennessee Press, Knoxville, pp. 63-98. cetes. In: MCLaughlin 0], McLaughlin EG, Lemke PA (eds), The Bruns TD, Szaro TM, 1992. Rate and mode differences between Mycota. Part B: Systematics and Evolution, vol. VII. Springer- nuclear and mitochondrial small-subunit rRNA genes in Verlag, Berlin, pp. 121-168. mushrooms. Molecular Biology and Evolution 9: 836-855. Hosaka K, Bates ST, Beever RE, Castellano MA, Colgan W, Bruns TD,Szaro TM, Gardes M, Cullings RW, Pan]], Taylor DL, Dominguez LS, Nouhra ER, Geml], Giachini A],Kenney SR, Horton TR, Kretzer A, Garbelotto M, Li Y, 1998. A sequence Simpson NB, Spatafora ]W, Trappe ]M, 2006. Molecular phy- database for the identification of ectomycorrhizal basidiomy- logenetics of the gomphoid-phalloid fungi with an establish- cetes by phylogenetic analysis. Molecular Ecology 7: 257-272. ment of the new subclass and two new Castellano MA, 1988. The taxonomy of the genus Hysterangium (Ba- orders. Mycologia 98: 949-959. sidiornycotina, Hysterangiaceae) with notes on its ecology. PhD Huelsenbeck ]P, Ronquist F, 2001. MrBayes: Bayesian inference of Dissertation, Oregon State University, Corvallis, OR, USA, 227 pp. phylogenetic trees. Bioinformatics 17: 754-755. Colgan W, Castellano MA, Spatafora JW, 1997. Systematics of the Humpert A], Muench EL, Giachini A], Castellano MA, Spatafora]W, Hysterangiaceae (abstract). Inoculum 48: 7. 2001. Molecular phylogenetics of Ram aria (Gomphales) and related Corner EJH, 1950. A monograph of Clavaria and allied genera. genera: evidence from nuclear large subunit and mitochondrial Annals of Botany Memoir 1: 1-740. small subunit rONA sequences. Mycologia 93: 465-477. Corner EJH, 1966. A monograph of the cantharelloid fungi. Annals Iulich W, 1981. Higher taxa of Basidiomycetes. Bibliotheca Myco- of Botany Memoir 2: 1-255. logica 85: 1-485. Corner EJH, 1969. Notes on cantharelloid fungi. Nova Hedwigia 18: Kishino H, Hasegawa M, 1989. Evaluation of the maximum like- 783-818. lihood estimate of the evolutionary tree topologies from DNA Corner EJH, 1970. Supplement to "A monograph of Clavaria and sequence data, and the branching order in hominoidea. Journal allied genera".Beihefte zur Nova Hedwigia 33: 1-299. of Molecular Evolution 29: 170-179. Corner EJH, 1972. Boletus in Malaysia. Botanic Gardens, Singapore, Lebel T, 1998. The sequestrate relatives of Russula in Australia and 263 pp. New Zealand. PhD Dissertation,Oregon State University, Cunningham GH, 1942. The Gasteromycetes of Australia and New Corvallis, OR, USA, 440 pp. Zealand. John McIndoe, Dunedin, 236 pp. Maddison WP, Maddison DR, 1992. MacClade (ver. 3). Sinauer As- Donk MA, 1961. Four new families of Hymenomycetes. Persoonia sociates, Sunderland, MA, USA. 1: 405-407. Maire R, 1902.Recherches cytologiques et taxonorniques sur les Ba- Donk MA, 1964. A conspectus of the families of Aphyllophorales. sidiomycetes. Bulletin de la Societe Mycologique de France 18: 1-209. Persoonia 3: 199-324. Maire R, 1914. La flare mycologique des forets de Cedres de l'At- Dring DM, 1973. Gasteromycetes. In: Ainsworth GC, Sparrow FK, las. Bulletin de la Societe Mycologique de France 30: 199-220. Sussman AS (eds), The Fungi, an Advanced Treatise, vol. 4B. Marr CD, Stuntz DE, 1973. Ramaria of western Washington. Bib- Academic Press, New York, pp. 451-478. liotheca Mycologica 38: 1-232. Eriksson J. 1954. Ramaricium n. gen., a corticoid member of the Masui K, 1926. A study of the of Abies firma, S. et Z., Ramaria group. Svensk Botanisk Tidskrift 48: 188-198. with special reference to its mycorrhizal fungus, Cantharellus Fiasson J-L, Petersen RH, Bouchez M-P, Arpin N, 1970. Contribu- floccosus, Schw. Memoirs of the College of Science, Kyoto Imperial tion biochimique ii. la connaissance taxonomique de certains University, Series B 2: 1-84. champignons cantharelloides et clavarioides. Revue Mycologi- Masui K, 1927. A study of the ectotrophic of woody que 34: 357-364. plants. Memoirs of the College of Science, Kyoto Imperial University, Fischer ED, 1933. Gastromyceteae 7a. In: Engler, Prantl (eds), Die Series B 3: 149-279. Natural Pflanzenfam, II edn., pp. 1-122. Miller Jr OK, Miller HH, 1988. Gasteromycetes: Morphological and Fitzpatrick HM, 1913. A comparative study of the development of Developmental Features with Keys to Orders, Families, and Genera. the fruit body in , Hysterangium, and Gautieria. An- Mad River Press, Eureka, 157 pp. nals of Mycology Sydowia 11: 119-149. Miller Jr OK, Watling R, 1987. Whence cometh the Agarics? A re- Giachini A], Spatafora ]W, Cazares E, Trappe ]M, 2001. Molecular appraisal. In: Petersen RH (ed.), Evolution in the Higher Basidio- phylogenetics of Gomphus and related genera inferred from nuclear mycetes: an International Symposium. University of Tennessee large and mitochondrial small subunits ribosomal DNA sequences. Press, Knoxville, pp. 435-448. Abstracts of the Third International Conference on Mycorrhi- Moncalvo J-M, Lutzoni FM, Rehner SA, Johnson 1. Vilgalys R, 2000. zas (ICOM Ill). Adelaide, p. 95. Phylogenetic relationships of agaricoid fungi based on nuclear Giachini AJ, 2004. Systematics of the Gomphales: the Genus large subunit ribosomal DNA sequences. Systematic Biology 49: Gomphus Pers. sensu lato. PhD Dissertation, Oregon State Uni- 278-305. versity, Corvallis, OR, USA, 446 pp. Mueller GM, Pine EM, 1994. DNA data provide evidence on the Griffiths RP, Castellano MA, Caldwell BA, 1991. Hyphal mats formed evolutionary relationships between mushrooms and false by two ectomycorrhizal fungi and their association with truffles. McIlvanea 11: 61-74. Douglas-fir seedlings: a case study. Plant and Soil 134: 255-259. Pagel M, 2003. Multistate v.0.8. Division of Zoology,School of An- Harrison KA, 1971. The evolutionary lines in the fungi with spines imal and Microbial Sciences, University of Reading, UK. supporting the hymenium. In: Petersen RH (ed), Evolution in the Petersen RH, 1968. Notes on cantharelloid fungi. I. Gomphus S.F. Higher Basidiomycetes: an International Symposium. University of Gray and some clues to the origin of the ramarioid fungi. The Tennessee Press,Knoxville, pp. 375-392. Journal of the Elisha Mitchell Scientific Society 84: 373-381. Petersen RH, 1971a. Interfamilial relationships in the c1avarioid Shimodaira H,Hasegawa M, 1999. Multiple comparisons of and cantharelloid fungi. In: Petersen RH (ed), Evolution in the log-likelihoods with applications to phylogenetic inference. Higher Basidiomycetes. An International Symposium. University of Molecular Biology and Evolution 16: 1114-1116. Tennessee Press, Knoxville, pp. 345-374. Singer R, 1947. Coscinoids and coscinocystidia in Lynderomyces Petersen RH, 1971b. The genera Gomphus and Gloeocantharellus in lateritius. Farlowia 3: 155-157. North America. Nova Hedwigia 21: 1-118. Singer R, 1949. The (mushrooms) in modern taxon- Petersen RH, 1973. Aphyllophorales II: the clavarioid and can- omy. Lilloa 22: 1-832. tharelloid Basidiomycetes. In: Ainsworth GC, Sussman AS Singer R, 1986. The Agaricales in Modern Taxonomy. Koeltz Scientific (eds), The Fungi, an Advanced Treatise. Academic Press, New Books, Koenigstein, 981 pp. York, pp. 351-367. Smith AH, 1973. Agaricales and related secotioid Gasteromycetes. Petersen RH, 1981. Ramaria subgenus Echinoramaria. Bibliotheca In: The Fungi IVB, a Taxonomic Review with Keys: Basidiomycetes Mycologica 79: 1-261. and Lower Fungi. Academic Press,London, pp. 421-450. Petersen RH, 1988. The Clavarioid Fungi of New Zealand. Science Swofford D, 2002. PAUP* 4.0blO. Sinauer Associates, Sunderland, Information Publication Centre, Wellington, 170 pp. MA, USA. Pine EM,Hibbett DS, Donoghue M], 1999. Phylogenetic relation- Thiers HD, 1984. The secotioid syndrome. Mycologia 76: 1-8. ships of cantharelloid and clavarioid Homobasidiomycetes Thiers HD, Trappe ]M, 1969. Studies in the genus Gastroboletus. based on mitochondrial and nuclear rDNA sequences. Brittonia 21: 244-254. Mycologia 91: 944-963. Villegas M, De Luna E, Cifuentes J, Estrada-Torres A, 1999. Phy- Roberts P, 1999. Clavarioid fungi from Korup National Park, logenetic studies in Gomphaceae sensu lato (Basidiomycetes). Cameroon. Kew Bulletin 54: 517-539. Mycotaxon 70: 127-147. Schild E, 1998. Die Gattung Ramaria: 4 neue Arten aus Italien und Zeller SM, Dodge CW, 1918. Gautieria in North America. Annals of Sardinien. Zeitschrift fur Mykologie 64: 53-66. the Missouri Botanical Garden 5: 133-142.